Reforming of hydrocarbon fuels to make hydrogen is well known in the art. Conventionally, hydrocarbons are reformed predominately in large-scale industrial facilities providing hydrogen for bulk storage and redistribution, or producing hydrogen as an on-line, upstream reagent for another large-scale chemical process. For the most part, these prior processes operate continuously and at steady-state conditions.
More recently, however, a strong interest has developed in providing hydrocarbon-reforming reactors integrated with an end use of the hydrogen. Also, there is a strong interest to develop a low-cost, small-scale source for hydrogen that can replace the need for storing hydrogen gas on site or on board. More particularly, a great interest has developed in providing reactors for producing hydrogen, which can be integrated with a fuel cell which uses hydrogen as a fuel source to generate electricity. Such hydrogen generator/fuel cell systems are being pursued for stationary uses such as providing electrical power to a stationary facility (home or business), for portable electric power uses, and for transportation.
The use of fuel cells, such as polymer electrolyte membrane fuel cells (PEM-FC), has been proposed for many applications, specifically including electrical vehicular power plants used to replace internal combustion engines. Electrochemical fuel cells convert fuel and oxidant to electricity and reaction product. Hydrogen is most commonly used as the fuel and is supplied to the fuel cell's anode. Oxygen (commonly as air) is the cell's oxidant and is supplied to the cell's cathode. The reaction product is water.
Efficiency and low emissions are two benefits of fuel cell systems. A system running near 40% efficiency will offer the opportunity to significantly reduce fuel consumption and CO2 production compared to conventional gasoline internal combustion engines. Perhaps more importantly, it has been shown that fuel cell systems, even when running with an onboard fuel processor, offer an opportunity to greatly reduce emissions of NOx, carbon monoxide, and hydrocarbons in automotive applications.
There are many technical requirements for reactors used in such applications, which are not required of traditional large or small-scale hydrogen generating reactors. For example, it is of particular interest to have such a system where the fuel cell can provide “power on demand.” Hence, hydrogen must be produced at required variable levels on demand. In other words, the hydrogen producing reactors must be sufficiently dynamic to follow the load. It is also of interest that such systems perform well upon start-up and shutdown cycling. In particular, it is desirable to have these integrated systems be stable through repeated on-off cycling including being ready to come back on-line in a relatively short time.
Another marked difference between proposed integrated systems and traditional reactors is that there must be sufficient processing in the integrated system itself, of the hydrocarbon feed stock so as to not only give a yield of hydrogen sufficient to meet the demand, but also to minimize byproducts of reaction including contaminants. In large-scale reactor systems, which produce enormous volumes and run continuously; space, weight, and cost of auxiliary systems is not so critical as in the integrated, smaller-scale reformers, especially those proposed for portable power or transportation applications. For example, carbon monoxide may be considered an undesirable reaction product on board a fuel cell powered automobile. However, in a steady state conventional process, the carbon monoxide can easily be handled by auxiliary separation systems, and may in fact be welcomed for its use in a synthesis gas to make acetic acid, dimethyl ether, and alcohols.
In short, the challenge for the smaller-scale, dynamic, integrated processors is the idea that what goes in the reformer, must come out at the same end as the desired hydrogen gas. Accordingly, processing has to be more complete and efficient, while it must also be cost effective, lightweight, and durable. The processing must be sufficient to reduce or eliminate species in the product gas which are harmful to the end use (for example, fuel cells) or other down stream components.
Another challenge exists for the proposed integrated systems with respect to the hydrocarbon feed stock. To be of maximum benefit, the proposed integrated systems should be able to use existing infrastructure fuels such as gasoline or diesel fuels. These fuels were not designed as a feed stock for generating hydrogen. Because of this, integrated systems are challenged to be able to handle the wide variety of hydrocarbons in the feed stock. For example, certain reforming byproducts such as olefins, benzene, methyl amide, and higher molecular weight aromatics can cause harm to catalysts used in reforming or purifying steps and may harm the fuel cell itself. Impurities in these fuels such as sulfur and chlorine can also be harmful to reactor catalysts and to the fuel cell.
It is also important to note, that a natural byproduct of hydrocarbon reforming is carbon monoxide. Carbon monoxide can poison proton exchange membrane fuel cells, even at very low concentrations of, for example, less than 100 ppm. Typical carbon monoxide levels exiting a fuel processing assembly (“FPA”) are about 2,000 to 5,000 ppm. This poses a problem for an integrated reactor system that is not faced by traditional reforming processes where significant carbon monoxide concentrations are either a useful co-product, or can be separated from the product gas without undue burden on the system economics as a whole.
Also, as noted above, integrated systems proposed to date are expected to transfer the total of the reformate to a fuel cell. Accordingly, techniques which separate carbon monoxide from hydrogen, such as pressure swing adsorption (“PSA”) or hydrogen permeable membrane separation, have the deficit of having to provide an alternate means for disposal or storage of the carbon monoxide. Both of the aforementioned techniques also suffer in efficiency as neither converts the carbon monoxide (in the presence of water) to maximize hydrogen production. PSA also suffers from high cost and space requirements, and presents a likely unacceptable parasitic power burden for portable power or transportation applications. At the same time, hydrogen permeable membranes are expensive, sensitive to fouling from impurities in the reformate, and reduce the total volume of hydrogen in the reformate stream.
One known method of reforming gaseous or liquid hydrocarbon fuels is by catalytic steam reforming. In this process a mixture of steam and the hydrocarbon fuel is exposed to a suitable catalyst at a high temperature. The catalyst used is typically nickel and the temperature is usually between about 700° C. and about 1000° C. In the case of methane, or natural gas, hydrogen is liberated in a catalytic steam reforming process according to the following overall reaction:CH4+H2O→CO+3H2  (1)
This reaction is highly endothermic and requires an external source of heat and a source for steam. Commercial steam reformers typically comprise externally heated, catalyst filled tubes and rarely have thermal efficiencies greater than about 60%.
Another conventional method of reforming a gaseous or liquid hydrocarbon fuel is partial oxidation (POx) reforming. In these processes a mixture of the hydrocarbon fuel and an oxygen containing gas are brought together within a POx chamber and subjected to an elevated temperature, preferably in the presence of a catalyst. The catalyst used is normally a noble metal or nickel and the high temperature is normally between about 700° C. and about 1200° C. for catalyzed reactions, and about 1200° C. to about 1700° C. for non-catalyzed reactions. In the case of methane, or natural gas, hydrogen is liberated in a POx chamber according to the following overall reaction:CH4+½O2→CO+2H2  (2)
This reaction is highly exothermic and once started generates sufficient heat to be self sustaining. That is, no external heat supply or steam supply is required. The catalytic partial oxidation reforming technique is simpler than the catalytic steam reforming technique, but is not as thermally efficient as catalytic steam reforming.
An additional known method of reforming a hydrocarbon fuel is by autothermal reforming, or “ATR”. An autothermal reformer uses a combination of steam reforming and partial oxidation reforming. Waste heat from the partial oxidation reforming reaction is used to heat the thermally steam reforming reaction. An autothermal reformer may in many cases be more efficient than either a catalytic steam reformer or a catalytic partial oxidation reformer. Again, using methane, or natural gas, as the hydrocarbon fuel, hydrogen is liberated according to the following overall reaction:CH4+yH2O+(1−y/2)O2 →CO2+(2+y)H2, where 0<y<2  (3)
Consideration of the standard enthalpies of formation shows that autothermal operation is theoretically achieved when y=1.115.
In addition to the reforming reactions discussed above it is usually necessary to consider the effects of another reaction occurring, the so called “water gas shift reaction.” Because the equilibrium of this reversible reaction is temperature (T) dependent, and at high temperatures carbon monoxide and water tend to be produced, the effects warrant consideration. In the water gas shift reaction the following overall reaction occurs:CO+H2O(g)→CO2+H2  (4)
More favorably, however, is that given equilibrium conversion at low temperatures carbon dioxide and hydrogen tend to be produced.
Typical reformers produce carbon dioxide and hydrogen, and consequently some carbon dioxide and hydrogen react to produce concentrations of carbon monoxide and water due to the reverse water gas shift reaction occurring in the reforming chamber. As mentioned previously, this is undesirable because the concentration of carbon monoxide must be either completely removed or at least reduced to a low concentration—i.e., less than about 100 ppm after the shift reaction—to avoid poisoning the anode of the PEM-FC. Carbon monoxide concentrations of more than 20 ppm reaching the PEM-FC can quickly poison the catalyst of the fuel cell 's anode. In a shift reactor, water (i.e., steam) is added to the hydrocarbon reformate/effluent exiting the reformer, in the presence of a suitable catalyst, to lower its temperature, and increase the steam to carbon ratio therein. The higher steam to carbon ratio serves to lower the carbon monoxide content of the reformate to less than 100 ppm according to the shift reaction (4) above. Ideally, the carbon monoxide concentration can be maintained below 1 ppm with the right shift catalyst, but the temperature required for this, about 100° C.–125° C., is too low for operation of current shift catalysts.
Advantageously, it is possible to recover some hydrogen at the same time by passing the product gases leaving the reformer, after cooling, into a shift reactor where a suitable catalyst promotes the carbon monoxide and water/steam to react to produce carbon dioxide and hydrogen. The shift reactor provides a convenient method of reducing the carbon monoxide concentration of the reformer product gases, while simultaneously improving the yield of hydrogen.
However, some carbon monoxide still survives the shift reaction. Depending upon such factors as reformate flow rate and steam injection rate, the carbon monoxide content of the gas exiting the shift reactor can be as low as 0.5 mol percent. Any residual hydrocarbon fuel is easily converted to carbon dioxide and hydrogen in the shift reactor. Hence, shift reactor effluent comprises not only hydrogen and carbon dioxide, but also water and some carbon monoxide.
The shift reaction is typically not enough to sufficiently reduce the carbon monoxide content of the reformate (i.e., below about 100 ppm). Therefore, it is necessary to further remove carbon monoxide from the hydrogen-rich reformate stream exiting the shift reactor, prior to supplying it to the fuel cell. It is known to further reduce the carbon monoxide content of hydrogen-rich reformate exiting a shift reactor by a so-called preferential oxidation (“PrOx”) reaction (also known as “selective oxidation”) effected in a suitable PrOx reactor. A PrOx reactor usually comprises a catalyst bed which promotes the preferential oxidation of carbon monoxide to carbon dioxide by air in the presence of the diatomic hydrogen, but without oxidizing substantial quantities of the H2 itself. The preferential oxidation reaction is as follows:CO+½O2→CO2  (5)
Desirably, the O2 required for the PrOx reaction will be no more than about two times the stoichiometric amount required to react the CO in the reformate. If the amount of O2 exceeds about two times the stoichiometric amount needed, excessive consumption of H2 results. On the other hand, if the amount of O2 is substantially less than about two times the stoichiometric amount needed, insufficient CO oxidation will occur. The PrOx process is described in a paper entitled “Preferential Oxidation of CO over Pt/ÿ-Al2O3 and Au/ÿ—Fe2O3: Reactor Design Calculations and Experimental Results” by M. J. Kahlich, et al. published in the Journal of New Materials for Electrochemical Systems, 1988 (pp. 39–46), and in U.S. Pat. No. 5,316,747 to Pow et al.
PrOx reactions may be either (1) adiabatic (i.e., where the temperature of the reformate (syngas) and the catalyst are allowed to rise during oxidation of the CO), or (2) isothermal (i.e., where the temperature of the reformate (syngas) and the catalyst are maintained substantially constant during oxidation of the CO). The adiabatic PrOx process is typically effected via a number of sequential stages which progressively reduce the CO content. Temperature control is important in both systems, because if the temperature rises too much, methanation, hydrogen oxidation, or a reverse shift reaction can occur. This reverse shift reaction produces more undesirable CO, while methanation and hydrogen oxidation negatively impact system efficiencies.
In either case, a controlled amount of O2 (e.g., as air) is mixed with the reformate exiting the shift reactor, and the mixture is passed through a suitable catalyst bed known to those skilled in the art. To control the air input, the CO concentration in the gas exiting the shift reactor is measured, and based thereon, the O2 concentration needed for the PrOx reaction is adjusted. However, effective real time CO sensors are not available and accordingly system response to CO concentration measurements is slow.
For the PrOx process to be most efficient in a dynamic system (i.e., where the flow rate and CO content of the hydrogen-rich reformate vary continuously in response to variations in the power demands on the fuel cell system), the amount of O2 (e.g., as air) supplied to the PrOx reactor must also vary on a real time basis in order to continuously maintain the desired oxygen-to-carbon monoxide concentration ratio for reaction (5) above.
Another challenge for dynamic operation is that the reformate at start-up contains too much carbon monoxide for conversion in the PrOx reactor and, therefore, is not suitable for use in a PEM-FC. One approach to this problem is to discharge this unsuitable reformate without benefit, and potentially to the detriment of the environment. The partially reformed material may contain unacceptable levels of hydrocarbons, carbon monoxide, sulfur, noxious oxides, and the like. It would be an advantage to provide a process which utilizes the waste reformate to assist in the preheating of unreformed fuel before its entry into the reforming chamber, while simultaneously converting the harmful constituents of the waste reformate to acceptable emissions.
A PEM-FC typically does not make use of 100% of the incoming hydrogen from the reformer/reactor. Therefore, anode gases—mostly unused hydrogen—are discharged from the fuel cell simultaneous with the input of hydrogen. It would be an advantage in the industry to make use of this combustible material to assist the preheating of unreformed hydrocarbon fuel or for steam generation. Systems already proposed employ so called “tail gas combusters” to burn off such fuel cell exhaust gases.
The present invention addresses the above problems and challenges and provides other advantages as will be understood by those in the art in view of the following specification and claims.